[PATCH] D50121: [PowerPC] Do not round values prior to converting to integer

Nemanja Ivanovic via Phabricator via llvm-commits llvm-commits at lists.llvm.org
Tue Jul 31 17:40:58 PDT 2018


nemanjai created this revision.
nemanjai added reviewers: cuviper, tstellar, hfinkel.
Herald added a subscriber: kbarton.

As pointed out in https://bugs.llvm.org/show_bug.cgi?id=38342, adding the FP_ROUND prior to converting from double precision to 4-byte integers causes loss of precision. This patch fixes that bug ensuring that we do not need to scalarize these conversions.


Repository:
  rL LLVM

https://reviews.llvm.org/D50121

Files:
  lib/Target/PowerPC/PPCISelLowering.cpp
  lib/Target/PowerPC/PPCInstrVSX.td
  test/CodeGen/PowerPC/build-vector-tests.ll

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D50121.158431.patch
Type: text/x-patch
Size: 37144 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20180801/3ebfc617/attachment.bin>


More information about the llvm-commits mailing list